1. Technical Field
The present invention generally relates to an apparatus and method for creating a three-dimensional (3D) figure and, more particularly, to an apparatus and method for automatically creating a 3D personalized figure suitable for 3D printing by detecting a face area and features for respective regions from face data acquired by heterogeneous sensors and by optimizing global/local transformation.
2. Description of the Related Art
Generally, methods of creating a 3D face model include methods of scanning static appearance information of an entity using an active sensor such as laser or pattern light; methods of creating a 3D face model either using multiview or stereo images in which conditions such as lighting, sensor geometric information, and viewpoints are previously set, or using a depth sensor or the like; and methods of a skilled expert manually creating a 3D model, etc.
For example, U.S. Patent Application Publication No. 2011-0148868 entitled “Apparatus and Method for Reconstructing Three-Dimensional Face Avatar through Stereo Vision and Face Detection” discloses a process for generating the face of a 3D avatar using a stereo vision system and a face detector.
Further, Korean Patent Application Publication No. 2012-0102275 entitled “Virtual Service System For Hair and Method thereof, and Device Supporting the Same” discloses technology for creating a 3D model by performing 3D rendering on the face and head of a user, and applying a specific hair object to the corresponding 3D model.
In this way, among the conventional methods of creating a 3D face model, the scanning method requires a post-processing method conducted by an expert, and the existing sensor-based method is disadvantageous in that good results can be obtained only in a preset condition. Further, since multiple cameras must be used, various disadvantages such as problems with synchronization between cameras, color coincidence, and geometric correction are present. Furthermore, a problem arises in that appearance data generated by scanning the overall region of a head including a face or by capturing only a part of the face is composed of a large number of unrefined dummy meshes having holes or overlapping, or includes only a part of face data, and thus it is not suitable to directly apply the appearance data to 3D printing.